Pilot Master Redundancy in Data Network Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data network bus systems are vulnerable to malfunctions when the pilot master or separate pilot signal generator fails, leading to system-wide failures.
Innovation Solution
Implementing multiple users as pilot-master-capable, where they check for external pilot signals and randomly determine a new pilot master based on checking time intervals, allowing any user to assume the role and preventing dependency on external controllers or predefined masters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single predefined user or separate pilot signal generator is used to generate the pilot signal, then the system structure is simple and easy to implement, but the system reliability deteriorates because a malfunction of the pilot master or pilot signal generator causes system-wide failure
Solution Approach 1:
The patent divides the pilot master function into multiple independent users (at least two pilot-master-capable users) instead of relying on a single pilot master. Each user can independently assume the pilot master role, segmenting the critical function across multiple nodes to eliminate single-point failure and improve system reliability.
Solution Approach 2:
The patent implements a random checking time interval mechanism where pilot-master-capable users monitor for external pilot signals during randomly assigned intervals. This beforehand cushioning ensures that if the current pilot master fails, another user is already prepared to detect the absence of pilot signals and take over, preventing system-wide failure.
2Reliability
If multiple users are made pilot-master-capable with random checking time intervals, then the system reliability improves through redundancy, but the complexity of determining pilot mastership increases
Solution Approach 1:
The patent introduces dynamic random checking time intervals for each pilot-master-capable user instead of static assignments. The random duration of checking time intervals allows the system to dynamically adapt to different operational conditions while maintaining simplicity through probabilistic rather than deterministic control logic.
Solution Approach 2:
Pilot-master-capable users autonomously determine their own checking time intervals and monitor for external pilot signals without requiring external controllers or complex centralized coordination. Each user independently executes the checking process and self-assigns pilot mastership based on who detects no external pilot signal first, simplifying the overall system architecture.
3Reliability
If users check for external pilot signals during random checking time intervals, then the system achieves automatic failover capability, but the time to establish a new pilot master increases due to random waiting periods
Solution Approach 1:
Pilot-master-capable users perform preliminary actions by continuously monitoring for external pilot signals during their assigned random checking time intervals even before a failure occurs. This preliminary monitoring ensures that when a pilot master failure happens, the system can immediately identify and establish a new pilot master without extended detection delays, reducing overall failover time.
Solution Approach 2:
The patent uses random checking time intervals as a parameter change strategy. By varying the checking interval duration randomly rather than using fixed intervals, the system optimizes the balance between detection speed and collision avoidance. The random parameter allows the system to adapt to different scenarios, achieving reliable failover while minimizing establishment time through statistical optimization.
Data Source
AI summary
A method is described for establishing one user from multiple users of a data network as a pilot master for emitting a pilot signal, to which the other users of the data network may synchronize themselves. In the related art, a pilot signal is typically generated by a separate pilot signal generator or a permanently predefined selected user. However, this has the disadvantage that if the pilot signal generator or the selected user malfunctions, synchronization and therefore also communication of the users with one another via the data network is no longer possible. In order to avoid this disadvantage, it is suggested that at least two of the users of the data network be implemented to be pilot-master-capable and one of them assume the pilot mastership according to a described method.